Transactions of the Korean Nuclear Society Virtual spring Meeting

May 13-14, 2021

BNCT dosimetry with thermoluminescent dosimeter

Sangmin Lee a, Junyoung Lee a, Geunsub Kim a, Sung-Joon Ye a*

a
Program in Biomedical Radiation Sciences, Department of Transdisciplinary Studies, Graduate School of
Convergence Science and Technology, Seoul National University, Seoul, Korea
*
Corresponding author: sye@snu.ac.kr

Abstract: A pair of thermoluminescent dosimeters made of lithium fluoride compounds could measure a thermal
neutron flux. The calibration factor (3.8 × 104 cm-2nC-1) evaluated in the previous study was verified for the
measurement of the thermal neutron flux in a research reactor. The conventional thermal neutron flux
measurement uses gold foil or gold wire, and the measured thermal neutron flux is used to evaluate the boron
dose, a dose contributing to the treatment in the boron neutron capture therapy. The use of a cadmium filter is
essential because the neutron cross-section of gold has a relatively low energy range showing a 1/v behavior.
However, thermoluminescent dosimeters made of lithium fluoride compounds have a much wider energy range
in which the cross-section of lithium is 1/v, and the tendency is similar to that of boron so that the dose of boron
is able to be evaluated more accurately than gold. To verify the measured calibration factor, a
research/educational reactor of Kyunghee University was used. When the thermal neutrons of the nuclear
reactor were measured using gold, it was measured as 2.17 × 10 4 cm-2W-1s-1, and the thermoluminescent
dosimeter measured as an average 2.29 × 10 4 cm-2W-1s-1 using a cadmium filter. The thermal neutron flux was
accurately measured. In conclusion, through this study, a more accurate method of evaluating boron dose using
a thermal fluorescence dosimeter was proposed.

1. Introduction
The Boron Neutron Capture Therapy (BNCT) is
binary radiation treatment modality. It delivers radiation
damage to the tumor cell selectively using 10B compound
and neutron beam. The 10B nuclei which have high
thermal neutron capture cross-section, release α particle
and 7Li ion in 10B(n, α)7Li reaction. These products have
high linear energy transfer characteristics and their
ranges are around single cell diameter (~10 μm) in the
tissue. The BNCT can selectively treat the tumor cell
with the targeted boron drug delivery [1, 2]. Therefore,
measuring thermal neutron flux should be done in air or
in phantom configuration is important to optimize
therapeutic effect at the treatment planning procedure.
The measurement of thermal neutron flux in the
neutron and gamma mixed radiation fields is commonly
done with gold foil activation method. However, this
method requires additional instruments such as a High
Purity Germanium (HPGe) detectors, liquid nitrogen
cooling system and lead shielding blocks. In this study,
we validated the calibration factor of thermoluminescent
dosimeters (TLD) To separate the neutron and gamma
induced signals of TLDs in the mixed radiation fields, we
used pairs of TLD-600 and TLD-700 and cadmium
sheets. The cadmium sheets were used to estimate the
thermal neutron induced signals from total neutron
induced signals.
The primary object of this study is to develop the
method to estimate the thermal neutron flux with TLDs.
The TLDs were irradiated at the research/educational
reactor and the measured thermal neutron flux was
compared with the gold activation analysis.
2. Materials and Methods
2.1 TLD reading
The 21 TLD-600 and 21 TLD-700 (Harshaw, USA)
chips were prepared for this study. Its diameter was 4.5
mm and the thickness was 0.6 mm. The TLDs used in
this study are lithium fluoride based materials
(LiF:Mg,Ti). TLD-600 was enriched by 6Li (95.6%)
whereas TLD-700 was enriched by 7Li(99.99%) [3]. Due
to the difference in neutron cross-section of 6Li and 7Li,
TLD-600 is sensitive to neutrons and TLD-700 is
insensitive to neutrons [4]. The TLDs were annealed
with an electric furnace at 400℃ for 1 hour followed by
100℃ for 2 hours to eliminate the residual signals before
the irradiation.
The readout of TLDs was done by using Harshaw TLD
reader (Model 3500). The TLDs were linearly heated
from 50 ℃ to 300 ℃ at 10 ℃/sec. The acquisition time
was 33.3 seconds.
The TL signal was converted in the unit of electric
charges by integrating the glow curves from channels 72
to 200. The thermal neutron induced TL signal was
separated by using the combination of TLD-600, TLD-
700 and cadmium sheets with their different interaction
probabilities of neutrons and gamma rays.
2.2 TLD validation
To evaluate whether the calculated TLD calibration
factor is valid, a research/educational reactor at
Kyunghee University, known as a thermal neutron
source, was used. The thermal neutron flux measured
through the applying calibration factors of TLDs was
compared with the thermal neutron flux using the
conventional gold wire activation method. At this time,
the thermal neutron flux is the Westcott fluence.
In this study, measurements were made using a Cd
sheet that absorbs thermal neutrons with high probability.
Therefore, if Cd sheet is placed in the existing beam port
 Transactions of the Korean Nuclear Society Virtual spring Meeting
May 13-14, 2021

in AGN-201K, it absorbs too many thermal neutrons and 4. Conclusions

lowers the criticality. Due to the insufficient margin of
the criticality of the educational/research reactor, the In this study, TLD-600, TLD-700 and cadmium sheets
neutron irradiation was conducted at the top of the vessel were used to validate the calibration factor. The
far from the core. established method enables to estimate thermal neutron
flux without complex instruments used in the
3. Results and discussion conventional method. The thermal neutron flux
measured by TLD showed an average difference of 5%
3.1 TLD validation with gold activation analysis.
For the further study, the uncertainty analysis about
The gold wire activation methods conducted in this the other features such as the errors of reader system
study are summarized in Table. 1. The thermal neutron (including a high voltage supplier, photomuliplier tube
flux measured by the gold wire activation method is 2.00 and heating system) and experimental setup errors.
× 105 cm-2s-1W-1. The self-shielding correction factor of
0.2 mm diameter gold wire was 0.928 by MCNP6 5. Acknowledgements
simulations. The corrected value was 2.17 × 10 5 cm-2s-
1 -1
W and it used as the standard for thermal neutron flux The authors thank to Prof. Myung-Hyun Kim (KHU)
measurement. for the assistance during the neutron irradiation with the
reactor for this study.
Irradiation Bare Cd covered
Gold mass REFERENCES
183.1 185.6
[mg]
Cooling time
[sec]
9289 1523 [1] W. A. Sauerwein, A. Wittig, R. Moss, Y. Nakagawa,
Measuring time
7212 7211
Neutron Capture Therapy, Springer, Berlin, 2012
[sec] [2] R. Moss, Critical review with and optimistic outlook on
Count 15916 677 Boron Neutron Capture Therapy (BNCT), Applied Radiation
Asat per nuclide 2.06 × 10-17 8.56 × 10-19
and Isotopes, Vol. 88, pp. 2-11, 2014.
[3] Model 3500 Manual TLD Reader With WinREMS,
Neutron flux (Westcott)
2.08 × 105 8.56 × 103 Operator's Manual, Thermo Scientific. 2010.
[cm-2s-1W-1]
Thermal neutron flux (Westcott)
2.00 × 105
[4] D.A. Brown, et al., ENDF/B-VIII. 0: the 8th major release
[cm-2s-1W-1] of the nuclear reaction data library with CIELO-project cross
Table 1. Summary of gold wire activation method in this sections, new standards and thermal scattering data, Nuclear
study. Data Sheets, 148, p. 1-142. 2018.
[5] T. Goorley, et al., Initial MCNP6 Release Overview,
Table. 2 shows the thermal neutron flux at the Nuclear Technology, 180, pp 298-315, 2012
irradiation site of the reactor by multiplying the TLD [6] ICRU Report 57, Journal of the International Commission
reading value obtained in the TLD validation setup by on Radiation Units and Measurements, Volume os29, Issue 2,
1, 1998
the calibration factor. In 4 measurements, when the
thermal neutron flux was normalized to the reactor power
and irradiation time, the average value was 2.285 × 10 5
cm-2W-1s-1. The thermal neutron flux measured by TLD
showed an average difference of 5%.

#1 (2W, #2 (2W, #3 (4W, #4 (4W,

Irradiation
0.5 h) 0.5 h) 0.5h) 0.5h)

Thermal neutron
induced signal 6720.67 6396.22 13873.72 13573.92
(nC)

Calibration factor
3.8 × 104
[cm-2 nC-1]

Thermal neutron 2.29 × 105 2.18 × 105 2.36 × 105 2.31 × 105
flux
(Wescott)
[cm s W-1]
-2 -1
Gold wire: 2.17 × 105

Relative difference
1.05 1.004 1.08 1.06
[ Flux TLD / Gold]

Table 2. The results of validation of TLDs at the reactor.